Cold email or cold DM

Use email when your buyer works somewhere their inbox is the job, and a DM when your buyer is a founder or marketer whose real conversations happen on a platform. The deciding factor is where the person already answers messages, not which channel performs better in aggregate, because the aggregate numbers describe different populations rather than different techniques.

The worse mistake than choosing wrong is running both badly, which is the most common outcome and the one worth most of this page.

What each channel is actually good at

Email

Reach and reliability. Nearly every business buyer has an address, addresses follow guessable patterns, and nothing stops you sending. It is the only channel that works for enterprise, for regulated industries, and for anyone senior at a large company.

Its weakness is the crowd. You are arriving in the most contested attention space in business, alongside everyone else who worked out the same thing.

Direct messages

Intimacy and speed. The inbox is emptier, the recipient can see who you are before deciding, and you learn within a day whether it worked.

Its weaknesses are reach and fragility. The audience is limited to people active on that platform, finding the right account is real work, and the platform can restrict you for behaviour that email providers would not notice.

Choosing, in one question

Where does this person have conversations they actually want to have?

A VP of marketing at a 2,000-person company lives in email and Slack. A DM from a stranger is an intrusion into a space they use for people they chose. Send email.

A founder of a twelve-person crypto project has an email address that goes to a shared inbox somebody checks weekly, and a Telegram they answer in four minutes. Send the DM.

Almost everything else follows from that. Company size is the crude proxy, and it works often enough to be useful: the larger the company, the more email wins.

What each costs you when it goes wrong

This is the asymmetry most comparisons leave out.

A bad email campaign costs you your sending domain. Spam complaints damage reputation, and a damaged domain affects everything sent from it, including invoices and replies to customers. Recoverable, slowly, and the usual answer is to send outbound from a separate domain so the blast radius is contained.

A bad DM campaign costs you an account, and possibly an audience you spent a year building. Restrictions are often silent, so you can be sending into nothing for days. Less recoverable than a domain, which is why outreach should run from an account you could lose.

Why running both badly is the common failure

The reasoning goes: email gets 3%, DMs get 15%, so do both and get more. What happens in practice is that the research budget per prospect gets halved, both channels start running on templates, and both underperform the one that would have been done properly.

The constraint is not channels. It is how many companies you can look at properly in a week, and that number does not change when you add a second inbox to send from.

Using both well

There is a version that works, and it is sequential rather than parallel.

Same prospect, one channel first, the other only after the first produced nothing and only if the second is plausibly a better fit for that specific person. Space them out by a couple of weeks. The same message arriving in two places on the same day does not read as thoroughness, it reads as a tool with a checkbox enabled.

And never reference the other message. "I also emailed you about this" tells the recipient they are in a system.

The short answer

Pick the channel your buyer answers on. Do that one properly, at a volume where every message contains something real. Add the second only when the first is working and you have the hours to research for it, not because a comparison table said the reply rate was higher.